Re: [SEGA Space Ship] Trying to repair, status update.

From: Pat Danis <patdanis_at_verizon.net>
Date: Tue Jan 26 2010 - 15:32:55 EST
Thanks for the updates.  I tried following your posting in French but found that my High School education did not provide enough of a background to translate the page so that I could fully understand it.  Fortunately Google translated the page in such a way that I am able to follow your progress.  Despite the better translations, I am still finding that the terminology and level of engineering complexity is still above my head.  However, I shall continue to follow your progress.  Keep posting your findings.  I am intrigued by your progress.

Pat Danis

Fabrice GIRARDOT wrote:

Hi all.

This mail is a status update regarding my SEGA Space Ship PCB.
This PCB is a Cinematronic CCPU clone which I'm trying to fix.

I decided to disable the clock generator (removed x-tal and
1 resistor) and to setup a manual clock generator.
I'm using a push button and a flip-flop to avoid rebound
problems.

Now, I'm able to do step-by step debugging of the PCB.
Since my "manual clock" is before H2, I can check PCB during
rising AND falling edges of the clock.

Of course this is not the fastest way to fix my PCB, but
I hope it'll produce good results.
I don't have an Exorciser, and I don't own a logic analyser yet
(but I wish I do).

I dumped PROM D14. It looks like it has 16 bad values for data
bit "DS0" (pin 6). It reads "0" while the dump I found says "1".
Bad values are for addresses 0 1 2 3 4 9 A B F 10 11 12 16 17
1A and 1E.
This signal drives the 3 selectors at L11/M11/N11.
I don't know if it's really a chip problem or SEGA specific.

I decided to try to continue to fix my PCB while waiting for
an other chip for replacement.

Firstly, data read at PROM adress 0x000 was not 49 E7 as expected
(I assume SEGA's PROMs are bit to bit equals to Cinematronic's,
which I did not verify yet).
My CCPU reads 09 E7. After checks, I have 1 PROM that has
either a selection problem or an output problem.
PROM "PR-10" should be selected only when reaching addresses
starting at 0x800, but it tries to set one data bit to LOW.
I temporarily removed it and now my CCPU reads 49 E7.
At least, now I know SEGA's PROMs start like Cine's for sure.

I checked that this instruction (LDJ #7E9) is correctly
executed. I can see this value beeing written to P13/R13
as expected.
Then several other instructions are correctly executed.
CCPU quickly reaches a "58" which is "JMP" at the previously
loaded J register value.
My CCPU jumps at 7ED which is 1 bit wrong.
Strange, because I did closely check the value "9" beeing written
to P13 and "7E" to R13.
Answer is a bad 74LS194 at P13. One output has a strange value
of 1.4 V for a LOW state, and moreover jumps to HIGH on the falling
edge of pin 11.

I'm quiet happy because my CCPU heart seems to be living.
Maybe my PR-10 PROM is not fully dead.
*IF* it has "only" a selection problem, I'll give a chance to
its other selection input pin. If this input is still working,
it'll let me wait for a new PROM.

To be (slowly) continued.


PS: WIP can be seen there :
http://www.gamoover.net/Forums/index.php?topic=20677.msg300572#msg300572


Regards,


No virus found in this incoming message. Checked by AVG - www.avg.com Version: 9.0.730 / Virus Database: 271.1.1/2646 - Release Date: 01/26/10 01:46:00
--------------------------------------------------------------------------- ** Unsubscribe, subscribe, or view the archives at http://www.vectorlist.org ** Please direct other questions, comments, or problems to chris@westnet.com Received on Tue Jan 26 15:33:22 2010

This archive was generated by hypermail 2.1.8 : Tue Jan 26 2010 - 18:50:02 EST